Predicting Cardiovascular Risk in Patients with Prostate Cancer Receiving Abiraterone or Enzalutamide by Using...

Area of Science:

• Oncology
• Cardiology
• Medical Informatics

Background:

• Metastatic prostate cancer (PCa) patients undergoing androgen receptor pathway inhibitors (ARPIs) face significant cardiovascular risks.
• Proactive identification of these cardiovascular risks before treatment initiation is crucial but challenging.

Purpose of the Study:

• To develop and validate a machine learning model for predicting major adverse cardiovascular events (MACEs) in PCa patients receiving ARPIs.
• To identify key predictive indicators of MACEs in this specific patient cohort.

Main Methods:

• A nationwide cohort study utilized data from Taiwan's National Health Insurance Research Database and Cancer Registry.
• A random survival forest (RSF) model, incorporating 16 variables, was developed using a training set (n=3318) and validated on a separate set (n=1421).
• The model predicted MACEs over a mean follow-up of 2.1 years.

Main Results:

• The RSF model identified five key predictors of MACEs: age (<65 or ≥75 years), history of heart failure, stroke, hypertension, and myocardial infarction.
• The model demonstrated strong predictive performance with an AUC of 85.1% in the training set and 85.5% in the validation set.
• A positive correlation was found between the number of identified risk factors and the incidence of MACEs.

Conclusions:

• A machine learning approach effectively identified five critical predictors of MACEs in metastatic prostate cancer patients treated with ARPIs.
• These findings underscore the necessity of comprehensive cardiovascular risk assessment and diligent monitoring for PCa patients initiating ARPI therapy.